Context in Practice:
The Quality Bar for
AI-generated Context
How Zhenni Hu's team at Mastercard enriched 30,000+ assets and saved 6,000+ hours — and what the quality bar takes inside a regulated enterprise.
TRUSTED BY 500+ AI-FORWARD ENTERPRISES
About This Session
Zhenni Hu leads data management at Mastercard, where her team spent years building the governance foundation — the data quality, lineage, and stewardship — behind how Mastercard's data is understood and used. Context Agents ran against that work to enrich 30,000+ assets and save over 6,000 hours.
Zhenni joins Nandini Tyagi from Atlan's Founder's Office to share how she rolled Context Agents out inside a regulated enterprise: why she started with a controlled scope, how she brought governance stakeholders along, and how her team's role shifted from drafting descriptions to reviewing and certifying them.
What You Will Learn
When AI-generated context actually works
Why AI-generated context is viable now, and how to tell the difference between enrichment grounded in how your business actually uses its data and enrichment that won't survive a domain expert's review.
Mastercard's approach
Why a controlled scope came first, how Zhenni brought her governance stakeholders along, and what it looks like to shift a team's role from drafting descriptions from scratch to reviewing and certifying them.
What changes for the business and the team
For the business: a data estate that AI agents can actually use. For the team: a shift from documentation work to the judgment work only they can do — and what that role becomes on the other side.
Who this is for
This session is for data and AI leaders who want to understand what shipping AI-generated context at enterprise scale actually requires — whether you're evaluating a rollout or already working through quality and governance at scale.
The only proven way to create context
- Watch Video
- Watch Video
- Watch Video
- Watch Video
